Development of Symmetry Evaluation Methods for Determining the Degree of Ventricle Symmetry in Ischemic Stroke
CSEF · 2017 Computational Systems & Analysis
Overview
Objectives/Goals The objective of the project was to identify and develop effective methods of symmetry evaluation in images and apply these techniques to an imaging-based stroke rehabilitation phase I clinical trial named Dose Optimization for Stroke Evaluation (DOSE), Also detecting changes using the shape of the lateral ventricles to understand adjustments in the rest of the brain. The symmetry of the left and right lateral ventricles is an effective indicator of spatially altered regions of the brain. Methods/Materials Open patients MRI image of the brain on MRIcron software, Segment the lateral ventricle, 3D fill the ventricle, Overlay pre and post images of the ventricle, save file as volume format (v.o.i.), Use Bayes' rule to incorporate into Matlab for contrast from CSF and image background, Provide image to Matlab to generate ventricle volume. Results 84 MR image studies were processed and the symmetry ratios were calculated. The ratios will be utilized to evaluate potential correlation with the enrolled subjects# MAL(Motor activity log) scores and WMFT(wolf motor function test) scores over time. We expect the symmetry ratios to correlate with improved arm and hand function from the two scores over time, Evaluated from the CSF volume in the lateral ventricle. Conclusions/Discussion This method of symmetry detection is novel because it provides a quantitative measure of the symmetry between non-uniform regions. The lateral ventricles have a non-uniform shape, which is unique between patients and contributes to the difficulty of applying current symmetry detection methods. This algorithm effectively and accurately provides meaningful quantitative data. The symmetry evaluation method and the projected correlations will be very valuable for diagnostic and therapeutic applications in stroke and other clinical trials.
Summary statement
To find a new way to treat stroke as accurately as possible, by providing a quantitative measure of the symmetry between non-uniform regions.
Help received
My mentors Ximing Wang and Kevin Ma provided the patients files assigning us to segment and to calculate the volume of the ventricles.
Competition history
- CSEF 2017
Resources
Related projects
CSEF · 2013
The Effect of Features on Dementia Detection Accuracy
ISEF · 2014
The SMART System: Stroke Management with Augmented Reality Technology
CSEF · 2018
Automated Feature Detection for Diagnosing Neurodegenerative Diseases
JSHS · 2023
Novel Imaging Approaches for the Quantification of Changes in Perivascular Space Volume and Morphology in Response to Transcranial Direct Current Stimulation
CSEF · 2014
An Inexpensive, Global, and Effective Means of Diagnosing Heart Disease via Computer Imaging
ISEF · 2021
Stroke Rehabilitation Using Artificial Intelligence K.A.K.I (Kinesthetic Augmented Kinematic Inference)
CSEF · 2017
A Novel Algorithm to Diagnose Medical Issues: Identifying Facial Paralysis via Convolutional Neural Networks
CSEF · 2013
IlluminaMed: Developing New AI for Use in a Biomedical Image Analysis Toolkit
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: California Science & Engineering Fair public projects